Self-healing risk assessment method and device for power distribution network, and computer readable storage medium

Abstract

The application discloses a power distribution network self-healing risk assessment method and device, and a computer readable storage medium. The method comprises the following steps: obtaining a plurality of risk events of a power distribution network; inputting the plurality of risk events into a self-healing model to process the risk events by using the self-healing model, and obtaining a plurality of self-healing actions for processing the risk events; obtaining execution states, response speeds and influence ranges of the plurality of self-healing actions; obtaining a fusion feature vector of the execution state, the response speed and the influence range in each self-healing action; and determining a risk level of each self-healing action according to the plurality of fusion feature vectors respectively. The application solves the technical problem that, in the related art, the angle of considering the power distribution network self-healing risk is not comprehensive enough, and the influence of the risk event on the entire system cannot be accurately reflected.

Description

Technical Field

[0001] This invention relates to the field of power grid risk assessment technology, and more specifically, to a method and apparatus for self-healing risk assessment of distribution networks, and a computer-readable storage medium. Background Technology

[0002] With the rapid development of power grid technology, the deepening of the smart grid concept, and the orderly development of distribution network construction, while the stability and security of the power grid are constantly improving, the scale of the power grid is also growing exponentially. Distribution automation is an important means to improve power supply reliability, enhance power supply capacity, and achieve high-efficiency and economical operation. Therefore, distribution automation is needed to achieve distribution network self-healing in order to effectively improve the stability of the power grid.

[0003] However, the modes of distribution network self-healing achieved by distribution automation are not the same, and the degree of impact on power supply reliability is also different. There are many factors for assessing the risk of distribution network self-healing, but most of them are analyzed from a single perspective or some perspectives. They do not assess the factors of distribution network self-healing risk level, the probability and severity of accidents, and thus cannot fully reflect the impact of accidents on the entire system.

[0004] There is currently no effective solution to the problem that the aforementioned technologies do not take into account all aspects when assessing the self-healing risks of distribution networks and cannot accurately reflect the impact of risk events on the entire system. Summary of the Invention

[0005] This invention provides a method and apparatus for assessing the self-healing risk of a distribution network, as well as a computer-readable storage medium, to at least solve the technical problem in related technologies that the assessment of the self-healing risk of a distribution network is not comprehensive enough and cannot accurately reflect the impact of risk events on the entire system.

[0006] According to one aspect of the present invention, a method for self-healing risk assessment of a distribution network is provided, comprising: acquiring multiple risk events of the distribution network, wherein the risk events are events that cause a fault in the distribution network; inputting the multiple risk events into a self-healing model to process the risk events using the self-healing model to obtain multiple self-healing actions for processing the risk events, wherein the self-healing model is trained using multiple sets of first training data through deep learning, each set of the multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event; acquiring the execution status, response speed, and impact range of the multiple self-healing actions, wherein the execution status refers to the incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range at which the risk event causes a fault in the distribution network; acquiring a fused feature vector of the execution status, response speed, and impact range of each self-healing action; and determining the risk level of each self-healing action based on the multiple fused feature vectors.

[0007] Optionally, obtaining the fused feature vector of the execution state, the response speed, and the influence range in each self-healing action includes: performing feature transformation on the execution state to obtain a first feature vector of the execution state; performing feature transformation on the response speed to obtain a second feature vector of the response speed; performing feature transformation on the influence range to obtain a third feature vector of the influence range; and fusing the first feature vector, the second feature vector, and the third feature vector to obtain the fused feature vector of the execution state, the response speed, and the influence range.

[0008] Optionally, the risk level of each self-healing action is determined based on multiple fusion feature vectors, including at least one of the following: determining the risk level of the self-healing action based on a self-healing risk library and the fusion feature vectors, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process the historical risk events, and historical fusion feature vectors of the historical self-healing actions; and processing the fusion feature vectors using a risk assessment model to obtain the risk level of the self-healing action, wherein the risk assessment model is trained using multiple sets of second training data through machine learning, and each set of the multiple sets of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

[0009] Optionally, before determining the risk level of the self-healing action based on the self-healing risk library and the fusion feature vector, the self-healing risk assessment method for the distribution network includes: acquiring multiple historical risk events of the distribution network within a historical time period; processing the multiple historical risk events using the self-healing model to obtain multiple historical self-healing actions for processing the multiple historical risk events; determining the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; and establishing the self-healing risk library based on the correspondence between the multiple historical risk events, the multiple historical self-healing actions, and the multiple historical fusion feature vectors.

[0010] Optionally, determining the risk level of the self-healing action based on the self-healing risk database and the fusion feature vector includes: comparing multiple fusion feature vectors with historical fusion feature vectors of historical self-healing actions in the self-healing risk database to obtain comparison results; determining multiple historical fusion feature vectors identical to the fusion feature vectors as target fusion feature vectors based on the comparison results; dividing multiple self-healing actions into multiple self-healing action sets based on the multiple target fusion feature vectors, and calculating the average of the target fusion feature vectors of multiple self-healing actions in each self-healing action set to obtain multiple fusion feature averages; determining multiple risk levels equal to the number of self-healing action sets; determining the correspondence between the self-healing action sets and the risk levels based on the fusion feature averages; and determining the risk level of multiple self-healing actions in each self-healing action set based on the correspondence.

[0011] Optionally, after determining the risk level of the self-healing action based on the fused feature vector, the self-healing risk assessment method for the distribution network further includes: sorting the risk levels of multiple self-healing actions in descending order to obtain a first descending order sorting result; determining the processing order for processing multiple risk events based on the first descending order sorting result, wherein the order of the risk levels in the first descending order sorting result is positively correlated with the processing order; and sequentially controlling the execution of the self-healing action for processing each risk event according to the processing order.

[0012] Optionally, after determining the risk level of the self-healing action based on the fused feature vector, the self-healing risk assessment method for the distribution network further includes: sorting the risk levels of multiple self-healing actions in descending order to obtain a second descending order sorting result; determining the execution order of multiple self-healing actions based on the second descending order sorting result, wherein the order of the risk levels in the second descending order sorting result is positively correlated with the execution order; and controlling multiple self-healing actions to process multiple risk events sequentially according to the execution order.

[0013] According to another aspect of the present invention, a self-healing risk assessment device for a distribution network is also provided, comprising: a first acquisition unit, configured to acquire multiple risk events of the distribution network, wherein the risk events are events that cause a fault in the distribution network; and a second acquisition unit, configured to input the multiple risk events into a self-healing model, so as to process the risk events using the self-healing model to obtain multiple self-healing actions for processing the risk events, wherein the self-healing model is trained using multiple sets of first training data through deep learning, and each set of the multiple sets of first training data includes: sample risk events, The sample self-healing action corresponding to the sample risk event; a third acquisition unit, used to acquire the execution status, response speed and impact range of multiple self-healing actions, wherein the execution status refers to the incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range of the distribution network failure caused by the risk event; a fourth acquisition unit, used to acquire the fused feature vector of the execution status, response speed and impact range of each self-healing action; a first determination unit, used to determine the risk level of each self-healing action based on the multiple fused feature vectors.

[0014] Optionally, the fourth acquisition unit includes: a first acquisition module, configured to perform feature transformation on the execution state to obtain a first feature vector of the execution state; a second acquisition module, configured to perform feature transformation on the response speed to obtain a second feature vector of the response speed; a third acquisition module, configured to perform feature transformation on the influence range to obtain a third feature vector of the influence range; and a fourth acquisition module, configured to fuse the first feature vector, the second feature vector, and the third feature vector to obtain the fused feature vector of the execution state, the response speed, and the influence range.

[0015] Optionally, the first determining unit includes at least one of the following: a first determining module, configured to determine the risk level of the self-healing action based on the self-healing risk library and the fusion feature vector, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process the historical risk events, and historical fusion feature vectors of the historical self-healing actions; and a fifth obtaining module, configured to process the fusion feature vector using a risk assessment model to obtain the risk level of the self-healing action, wherein the risk assessment model is trained using multiple sets of second training data through machine learning, and each set of the multiple sets of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

[0016] Optionally, the self-healing risk assessment device for the distribution network includes: a sixth acquisition module, used to acquire multiple historical risk events of the distribution network within a historical time period before determining the risk level of the self-healing action based on the self-healing risk library and the fusion feature vector; a seventh acquisition module, used to process the multiple historical risk events using the self-healing model to obtain multiple historical self-healing actions that process the multiple historical risk events; a second determination module, used to determine the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; and an establishment module, used to establish the self-healing risk library based on the correspondence between the multiple historical risk events, the multiple historical self-healing actions, and the multiple historical fusion feature vectors.

[0017] Optionally, the first determining module includes: a first acquiring submodule, configured to compare multiple fusion feature vectors with historical fusion feature vectors of historical self-healing actions in the self-healing risk database to obtain a comparison result; a first determining submodule, configured to determine multiple historical fusion feature vectors identical to the fusion feature vectors as target fusion feature vectors based on the comparison result; a second acquiring submodule, configured to divide multiple self-healing actions into multiple self-healing action sets based on the multiple target fusion feature vectors, and calculate the mean of the target fusion feature vectors of multiple self-healing actions in each self-healing action set to obtain multiple fusion feature mean values; a second determining submodule, configured to determine multiple risk levels equal to the number of self-healing action sets; a third determining submodule, configured to determine the correspondence between the self-healing action sets and the risk levels based on the fusion feature mean values; and a fourth determining submodule, configured to determine the risk level of multiple self-healing actions in each self-healing action set based on the correspondence.

[0018] Optionally, the self-healing risk assessment device for the power distribution network further includes: a fifth acquisition unit, configured to sort the risk levels of multiple self-healing actions in descending order after determining the risk level of the self-healing action based on the fused feature vector, to obtain a first descending order sorting result; a second determination unit, configured to determine the processing order for processing multiple risk events based on the first descending order sorting result, wherein the order of the risk levels in the first descending order sorting result is positively correlated with the processing order; and a first control unit, configured to sequentially control the execution of the self-healing action for processing each risk event according to the processing order.

[0019] Optionally, the self-healing risk assessment device for the power distribution network further includes: a sixth acquisition unit, configured to sort the risk levels of multiple self-healing actions in descending order after determining the risk level of the self-healing action based on the fused feature vector, to obtain a second descending order sorting result; a third determination unit, configured to determine the execution order of multiple self-healing actions based on the second descending order sorting result, wherein the order of the risk levels in the second descending order sorting result is positively correlated with the execution order; and a second control unit, configured to control the multiple self-healing actions to process the multiple risk events sequentially according to the execution order.

[0020] According to another aspect of the present invention, a self-healing risk assessment system for a distribution network is also provided, wherein the self-healing risk assessment system for a distribution network uses any of the above-described methods for assessing the self-healing risk of a distribution network.

[0021] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes any of the above-described self-healing risk assessment methods for power distribution networks.

[0022] According to another aspect of the present invention, a processor is also provided, the processor being configured to run a program, wherein the program, when running, executes any of the above-described methods for self-healing risk assessment of power distribution networks.

[0023] According to another aspect of the present invention, a computer program product is also provided, including computer instructions, which, when executed by a processor, perform any of the above-described self-healing risk assessment methods for power distribution networks.

[0024] In this embodiment of the invention, multiple risk events of the distribution network are obtained, wherein a risk event is an event that causes a fault in the distribution network; the multiple risk events are input into a self-healing model to process the risk events, thereby obtaining multiple self-healing actions for processing the risk events; wherein the self-healing model is trained using multiple sets of first training data through deep learning, and each set of the multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event; the execution status, response speed, and impact range of the multiple self-healing actions are obtained, wherein the execution status refers to the incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range within which the risk event causes a fault in the distribution network; the fused feature vector of the execution status, response speed, and impact range of each self-healing action is obtained; and the risk level of each self-healing action is determined based on the multiple fused feature vectors. The above technical solutions achieve the goal of converting the execution status, response speed, and impact range of self-healing actions to resolve risk events into feature vectors, fusing them, and analyzing the risks of self-healing actions based on the fused feature vectors. This achieves the technical effect of assessing the self-healing risks of the distribution network from multiple perspectives, improving the accuracy of the self-healing risk assessment of the distribution network, and thus solving the technical problem in related technologies that the perspectives considered when assessing the self-healing risks of the distribution network are not comprehensive enough and cannot accurately reflect the impact of risk events on the entire system. Attached Figure Description

[0025] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0026] Figure 1 This is a hardware structure block diagram of a mobile terminal for a self-healing risk assessment method for a power distribution network according to an embodiment of the present invention.

[0027] Figure 2 This is a flowchart of a self-healing risk assessment method for a distribution network according to an embodiment of the present invention;

[0028] Figure 3 This is a schematic diagram of a self-healing risk assessment system for a power distribution network according to an embodiment of the present invention;

[0029] Figure 4 This is a schematic diagram of a self-healing risk assessment device for a power distribution network according to an embodiment of the present invention. Detailed Implementation

[0030] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0031] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0032] As described in the background section, related technologies do not consider all aspects when assessing the self-healing risks of distribution networks, and cannot accurately reflect the impact of risk events on the entire system. To address these shortcomings, embodiments of the present invention provide a method and apparatus for assessing the self-healing risks of distribution networks, as well as a computer-readable storage medium.

[0033] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0034] The methods and embodiments provided in this invention can be executed on a mobile terminal, a computer terminal, or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 This is a hardware structure block diagram of a mobile terminal for a self-healing risk assessment method for a power distribution network according to an embodiment of the present invention. Figure 1 As shown, a mobile terminal may include one or more ( Figure 1 Only one is shown in the diagram. A processor 102 (which may include, but is not limited to, a microprocessor MCU or a programmable logic device FPGA, etc.) and a memory 104 for storing data are also shown. The mobile terminal may further include a transmission device 106 for communication functions and an input / output device 108. Those skilled in the art will understand that... Figure 1 The structure shown is for illustrative purposes only and does not limit the structure of the mobile terminal described above. For example, the mobile terminal may also include components that are more... Figure 1The more or fewer components shown, or having the same Figure 1 The different configurations shown.

[0035] The memory 104 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the self-healing risk assessment method for power distribution networks in this embodiment of the invention. The processor 102 executes various functional applications and data processing by running the computer programs stored in the memory 104, thereby implementing the above-described method. The memory 104 may include high-speed random access memory and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 104 may further include memory remotely located relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof. The transmission device 106 is used to receive or send data via a network. Specific examples of the aforementioned networks may include wireless networks provided by the mobile terminal's communication provider. In one example, the transmission device 106 includes a Network Interface Controller (NIC), which can be connected to other network devices via a base station to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0036] According to an embodiment of the present invention, a method embodiment for self-healing risk assessment of a power distribution network is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0037] Figure 2 This is a flowchart of a self-healing risk assessment method for a distribution network according to an embodiment of the present invention, such as... Figure 2 As shown, the method includes the following steps:

[0038] Step S202: Obtain multiple risk events of the distribution network, wherein a risk event is an event that causes a fault in the distribution network.

[0039] Optionally, the aforementioned risk events may include, but are not limited to, various abnormal situations that may affect the normal operation of the distribution network, such as equipment failure (e.g., transformer failure, circuit breaker failure), line failure (e.g., short circuit, grounding failure), natural disasters (e.g., lightning strike, storm), human error (e.g., misoperation, illegal access), and load changes (e.g., starting or stopping of large equipment).

[0040] In this embodiment, when risk events exist in the distribution network, these risk events can be obtained first using a self-healing risk assessment system, and then processed.

[0041] It should be noted that using a self-healing risk assessment system to handle risk events in the distribution network is a dynamic process that allows for real-time monitoring and processing of the distribution network.

[0042] The following is combined Figure 3 The embodiments of the present invention will be described in detail below. Figure 3 This is a schematic diagram of a self-healing risk assessment system for a distribution network according to an embodiment of the present invention; as shown. Figure 3 As shown, the self-healing risk assessment system includes the following four modules: 1) Self-healing analysis module: used to analyze risk events and obtain self-healing actions to handle risk events using a self-healing model; 2) Self-healing action assessment module: used to conduct a preliminary assessment of the risk of self-healing actions; 3) Risk index calculation module: used to calculate the risk index (i.e., risk level) of self-healing actions; 4) Risk event ranking module: used to rank the priority of handling risk events according to the risk index of self-healing actions.

[0043] Step S204: Input multiple risk events into the self-healing model to process the risk events and obtain multiple self-healing actions for processing the risk events. The self-healing model is trained using multiple sets of first training data through deep learning. Each set of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event.

[0044] In this embodiment, risk events are input into the self-healing model of the self-healing analysis module. By analyzing the risk events through the self-healing model, self-healing actions to resolve the risk events can be obtained. The self-healing model here is specifically a deep learning network, which can be trained by a large number of historical risk events (i.e., sample risk events) and their corresponding historical self-healing actions (i.e. sample self-healing actions) to obtain the self-healing model.

[0045] Step S206: Obtain the execution status, response speed, and impact range of multiple self-healing actions. The execution status refers to the degree of incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the scope of the distribution network failure caused by the risk event.

[0046] In this embodiment, the self-healing actions analyzed by the self-healing analysis module can be input into the self-healing action evaluation module to conduct a preliminary assessment of the risks of the self-healing actions.

[0047] Specifically, data mining can be used to process self-healing actions to obtain their execution status, response speed, and impact range. The execution status indicates the degree of incompleteness of the self-healing action, the response speed indicates how fast the power distribution system executes the self-healing action, and the impact range indicates the extent of the impact of the risk event resolved by the self-healing action on the power distribution system after the self-healing action is completed.

[0048] Step S208: Obtain the fused feature vector of execution state, response speed and influence range in each self-healing action.

[0049] In this embodiment, the execution state, response speed, and influence range of the self-healing action can all be converted into feature vectors and fused to obtain the fused feature vector of the self-healing action.

[0050] According to the above embodiments of the present invention, in step S208, obtaining the fused feature vector of the execution state, the response speed, and the influence range in each self-healing action includes: performing feature transformation on the execution state to obtain a first feature vector of the execution state; performing feature transformation on the response speed to obtain a second feature vector of the response speed; performing feature transformation on the influence range to obtain a third feature vector of the influence range; and fusing the first feature vector, the second feature vector, and the third feature vector to obtain the fused feature vector of the execution state, the response speed, and the influence range.

[0051] Specifically, the execution state, response speed, and impact range of the self-healing action can all be converted into feature vectors. These feature vectors are then normalized to obtain a first feature vector representing the execution state, a second feature vector representing the response speed, and a third feature vector representing the impact range. The first, second, and third feature vectors are then fused to obtain a fused feature vector for the self-healing action. The larger the eigenvalue of the first feature vector, the higher the incompleteness of the self-healing action. A higher incompleteness indicates a longer recovery time for the power distribution system, which can easily lead to a decrease in the power supply reliability of the power distribution system, thus indicating a greater risk associated with the self-healing action. The larger the eigenvalue of the second eigenvector, the slower the response speed of the power distribution system in performing self-healing actions. A slower response speed means a longer recovery time, which can easily lead to a decrease in the power supply reliability of the power distribution system, indicating a greater risk of the self-healing action. The larger the eigenvalue of the third eigenvector, the larger the impact range on the power distribution system. A larger impact range means a greater likelihood of a decrease in the power supply reliability of the power distribution system, indicating a greater risk of the self-healing action. Therefore, the larger the eigenvalue of the fused eigenvector obtained from the first, second, and third eigenvectors, the greater the risk of the self-healing action.

[0052] Step S210: Determine the risk level of each self-healing action based on multiple fused feature vectors.

[0053] In this embodiment, the fusion feature vector of the self-healing action obtained by the self-healing action evaluation module can be input into the risk index calculation module to calculate the risk index of the self-healing action. That is, the risk level of the self-healing action can be determined by analyzing the risk index calculation module.

[0054] According to the above embodiments of the present invention, in step S210, determining the risk level of each self-healing action based on the plurality of fusion feature vectors includes at least one of the following: determining the risk level of the self-healing action based on a self-healing risk library and fusion feature vectors, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process historical risk events, and historical fusion feature vectors of historical self-healing actions; processing the fusion feature vectors using a risk assessment model to obtain the risk level of the self-healing action, wherein the risk assessment model is trained using multiple sets of second training data through machine learning, and each set of multiple sets of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

[0055] In this embodiment, the risk index calculation module can determine the risk level of a self-healing action based on the self-healing risk library; alternatively, it can use the risk assessment model in the risk index calculation module to analyze the self-healing action and obtain its risk level. Of course, the risk assessment model here is also based on the historical fusion feature vector (i.e., sample fusion feature vector) of the self-healing risk library to calculate the risk index of historical self-healing actions and obtain the risk index of historical self-healing actions (i.e., sample risk index); and the deep learning network is trained based on the historical self-healing actions and their risk indices.

[0056] According to the above embodiments of the present invention, before step S210, that is, before determining the risk level of the self-healing action based on the self-healing risk library and the fusion feature vector, the self-healing risk assessment method of the distribution network includes: acquiring multiple historical risk events of the distribution network within a historical time period; processing the multiple historical risk events using a self-healing model to obtain multiple historical self-healing actions for processing the multiple historical risk events; determining the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; and establishing a self-healing risk library based on the correspondence between the multiple historical risk events, the multiple historical self-healing actions, and the multiple historical fusion feature vectors.

[0057] Specifically, a self-healing risk database can be established based on the effects of self-healing actions and their impact on the power distribution system, and the evaluation results of historical self-healing actions. This involves: acquiring historical risk events and historical self-healing actions that resolved them; processing these historical self-healing actions through data mining to obtain their execution status, response speed, and impact range; converting these parameters into feature vectors and normalizing them to obtain a first feature vector representing the execution status, a second feature vector representing the response speed, and a third feature vector representing the impact range; fusing these three feature vectors to obtain a fused feature vector of historical self-healing actions; and establishing the self-healing risk database by combining historical risk events, historical self-healing actions that resolved them, and the fused feature vector of historical self-healing actions.

[0058] In the above embodiments of the present invention, determining the risk level of a self-healing action based on a self-healing risk database and fusion feature vectors includes: comparing multiple fusion feature vectors with historical fusion feature vectors of historical self-healing actions in the self-healing risk database to obtain comparison results; determining multiple historical fusion feature vectors identical to the fusion feature vectors as target fusion feature vectors based on the comparison results; dividing multiple self-healing actions into multiple self-healing action sets based on the multiple target fusion feature vectors, and calculating the average of the target fusion feature vectors of multiple self-healing actions in each self-healing action set to obtain multiple fusion feature averages; determining multiple risk levels equal to the number of self-healing action sets; determining the correspondence between self-healing action sets and risk levels based on the fusion feature averages; and determining the risk level of multiple self-healing actions in each self-healing action set based on the correspondence.

[0059] Specifically, the process can begin by comparing the fusion feature vector obtained through the self-healing action assessment module with the fusion feature vectors of self-healing actions in the self-healing risk database. Based on the comparison results, multiple fusion feature vectors identical to the fusion feature vectors are identified as target fusion feature vectors. Then, self-healing actions are classified according to these target fusion feature vectors to obtain a set of self-healing actions. Based on the target fusion feature vectors, a set feature vector is calculated for each set of self-healing actions. Based on the set feature vector, the risk index of each self-healing action included in the set is obtained. Each set of self-healing actions includes multiple self-healing actions. The calculation of the set feature vector for each set of self-healing actions is based on the target fusion feature vectors of the multiple self-healing actions included in the set. The mean of the target fusion feature vectors of self-healing actions is calculated to obtain the set feature vector of the self-healing action set. Since the larger the eigenvalue of the fusion feature vector of a self-healing action, the greater the risk of that self-healing action, the larger the eigenvalue of the set feature vector obtained from the fusion feature vector of the self-healing actions, the greater the risk of the self-healing action set. Classifying self-healing actions to obtain a set of self-healing actions allows self-healing actions with similar risks to be grouped together, which facilitates the subsequent definition of risk indicators. The number of risk indicator levels corresponds to the number of self-healing action sets. For example, if there are three self-healing action sets, the risk indicator levels can be divided into high risk, medium risk, and low risk. The specific risk indicator levels can be determined and designed according to actual needs.

[0060] It should be noted that the process of analyzing and determining the self-healing action is also the process of training the risk assessment model. In the specific analysis process, the above methods can be used directly to analyze and determine the risk level of the self-healing action, and the risk assessment model can be continuously trained and optimized in the process. Alternatively, after the risk assessment model matures, it can be used directly to analyze and determine the risk level of the self-healing action, and the risk assessment model can be continuously optimized in the process of use.

[0061] In an optional embodiment of the present invention, after determining the risk level of the self-healing action based on the fused feature vector, the self-healing risk assessment method of the power distribution network further includes: sorting the risk levels of multiple self-healing actions in descending order to obtain a first descending order sorting result; determining the processing order for processing multiple risk events based on the first descending order sorting result, wherein the order of risk levels in the first descending order sorting result is positively correlated with the processing order; and controlling the execution of the self-healing action for processing each risk event in sequence according to the processing order.

[0062] Specifically, the risk event sorting module can be used to sort the risk levels of self-healing actions obtained from the risk indicator calculation module in real time to determine the processing order of risk events.

[0063] In another optional embodiment of the present invention, after determining the risk level of the self-healing action based on the fused feature vector, the self-healing risk assessment method of the distribution network further includes: sorting the risk levels of multiple self-healing actions in descending order to obtain a second descending order sorting result; determining the execution order of multiple self-healing actions based on the second descending order sorting result, wherein the order of risk levels in the second descending order sorting result is positively correlated with the execution order; and controlling multiple self-healing actions to process multiple risk events sequentially according to the execution order.

[0064] Specifically, the risk event sorting module can be used to sort the execution order of self-healing actions in real time based on the risk level of the self-healing actions obtained from the risk indicator calculation module.

[0065] It should be noted that sorting the order of handling risk events or the order of executing self-healing actions is equivalent to sorting the priority of handling risk events. The higher the risk level of the self-healing action, the higher the degree of danger of the corresponding risk event, and the more priority it needs to be handled. In other words, it is necessary to control the execution of self-healing actions that can resolve the risk event.

[0066] The purpose of prioritizing risk events is to rank them according to the risk level of their self-healing actions, enabling power distribution system maintenance personnel to better understand the various risks facing the system and develop targeted response strategies, prioritizing high-risk events. Through this prioritization, personnel can quickly identify the most urgent and risky issues, addressing high-risk events first, thereby improving the safety and stability of the power distribution system and reducing the risk of potential failures or losses. Furthermore, the prioritization helps personnel allocate resources more efficiently, improving work efficiency and ensuring timely and effective responses when the system faces risks.

[0067] As described above, the technical solution provided by the above embodiments of the present invention can obtain multiple risk events in the distribution network, wherein a risk event is an event that causes a fault in the distribution network; multiple risk events are input into a self-healing model to process the risk events using the self-healing model, resulting in multiple self-healing actions for processing the risk events. The self-healing model is trained using multiple sets of first training data through deep learning, and each set of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event; the execution status, response speed, and impact range of multiple self-healing actions are obtained, wherein the execution status refers to the... The terms "incompleteness," "response speed" (referring to the speed at which the distribution network executes self-healing actions), and "impact range" (referring to the extent to which a risk event causes a fault in the distribution network) are defined. A fusion feature vector is obtained for the execution status, response speed, and impact range of each self-healing action. The risk level of each self-healing action is determined based on these fusion feature vectors. This achieves the technical effect of assessing the self-healing risk of the distribution network from multiple perspectives, improving the accuracy of the self-healing risk assessment.

[0068] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem in the related art that the perspectives considered when assessing the self-healing risk of the distribution network are not comprehensive enough, and the impact of risk events on the entire system cannot be accurately reflected.

[0069] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0070] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0071] According to embodiments of the present invention, a self-healing risk assessment device for a distribution network is also provided for implementing the above-described self-healing risk assessment method for a distribution network. Figure 4 This is a schematic diagram of a self-healing risk assessment device for a distribution network according to an embodiment of the present invention, as shown below. Figure 4 As shown, the device includes: a first acquisition unit 41, a second acquisition unit 43, a third acquisition unit 45, a fourth acquisition unit 47, and a first determination unit 49. The self-healing risk assessment device for this power distribution network will be described in detail below.

[0072] The first acquisition unit 41 is used to acquire multiple risk events of the distribution network, wherein the risk event is an event that causes a fault in the distribution network.

[0073] The second acquisition unit 43 is used to input multiple risk events into the self-healing model so as to process the risk events using the self-healing model and obtain multiple self-healing actions to process the risk events. The self-healing model is trained using multiple sets of first training data through deep learning. Each set of multiple sets of first training data includes: sample risk events and sample self-healing actions corresponding to the sample risk events.

[0074] The third acquisition unit 45 is used to acquire the execution status, response speed and impact range of multiple self-healing actions. The execution status refers to the degree of incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range of the distribution network failure caused by the risk event.

[0075] The fourth acquisition unit 47 is used to acquire the fused feature vector of execution state, response speed and influence range in each self-healing action.

[0076] The first determining unit 49 is used to determine the risk level of each self-healing action based on multiple fusion feature vectors.

[0077] It should be noted that the first acquisition unit 41, the second acquisition unit 43, the third acquisition unit 45, the fourth acquisition unit 47 and the first determination unit 49 mentioned above correspond to steps S202 to S210 in the above embodiments. The five units and the corresponding steps implement the same instances and application scenarios, but are not limited to the content disclosed in the above embodiments.

[0078] As can be seen from the above, in the scheme described in the above embodiments of the present invention, multiple risk events of the distribution network can be acquired first using the first acquisition unit, wherein the risk event is an event that causes a fault in the distribution network; then, the multiple risk events are input into the self-healing model using the second acquisition unit, so as to process the risk events using the self-healing model and obtain multiple self-healing actions for processing the risk events, wherein the self-healing model is trained using multiple sets of first training data through deep learning, and each set of multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event; then, the execution status, response speed and impact range of the multiple self-healing actions are acquired using the third acquisition unit, wherein the execution status refers to the self-healing action The degree of incompleteness of the action, the response speed refers to the speed at which the distribution network executes self-healing actions, and the scope of impact refers to the range of distribution network failures caused by the risk event. Then, the fourth acquisition unit is used to obtain the fused feature vector of the execution status, response speed, and scope of impact of each self-healing action. Finally, the first determination unit is used to determine the risk level of each self-healing action based on multiple fused feature vectors. This achieves the technical effect of assessing the self-healing risk of the distribution network from multiple perspectives by converting the execution status, response speed, and scope of the self-healing action to resolve the risk event into feature vectors and fusing them, so as to analyze the risk of the self-healing action based on the fused feature vector. This improves the accuracy of the self-healing risk assessment of the distribution network.

[0079] Therefore, the technical solution provided by the above embodiments of the present invention solves the technical problem in the related art that the perspectives considered when assessing the self-healing risk of the distribution network are not comprehensive enough, and the impact of risk events on the entire system cannot be accurately reflected.

[0080] Optionally, the fourth acquisition unit includes: a first acquisition module, used to perform feature transformation on the execution state to obtain a first feature vector of the execution state; a second acquisition module, used to perform feature transformation on the response speed to obtain a second feature vector of the response speed; a third acquisition module, used to perform feature transformation on the influence range to obtain a third feature vector of the influence range; and a fourth acquisition module, used to fuse the first feature vector, the second feature vector, and the third feature vector to obtain a fused feature vector of the execution state, response speed, and influence range.

[0081] Optionally, the first determining unit includes at least one of the following: a first determining module, used to determine the risk level of a self-healing action based on a self-healing risk library and a fusion feature vector, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process historical risk events, and historical fusion feature vectors of historical self-healing actions; and a fifth obtaining module, used to process the fusion feature vector using a risk assessment model to obtain the risk level of the self-healing action, wherein the risk assessment model is trained using multiple sets of second training data through machine learning, and each set of multiple sets of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

[0082] Optionally, the self-healing risk assessment device for the distribution network includes: a sixth acquisition module, used to acquire multiple historical risk events of the distribution network within a historical time period before determining the risk level of the self-healing action based on the self-healing risk library and fusion feature vectors; a seventh acquisition module, used to process the multiple historical risk events using a self-healing model to obtain multiple historical self-healing actions for processing the multiple historical risk events; a second determination module, used to determine the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; and an establishment module, used to establish a self-healing risk library based on the correspondence between the multiple historical risk events, the multiple historical self-healing actions, and the multiple historical fusion feature vectors.

[0083] Optionally, the first determining module includes: a first acquiring submodule, used to compare multiple fusion feature vectors with historical fusion feature vectors of historical self-healing actions in the self-healing risk database to obtain a comparison result; a first determining submodule, used to determine multiple historical fusion feature vectors that are identical to the fusion feature vectors as target fusion feature vectors based on the comparison result; a second acquiring submodule, used to divide multiple self-healing actions into multiple self-healing action sets based on the multiple target fusion feature vectors, and to calculate the mean of the target fusion feature vectors of multiple self-healing actions in each self-healing action set to obtain multiple fusion feature mean values; a second determining submodule, used to determine multiple risk levels that are the same number as the number of self-healing action sets; a third determining submodule, used to determine the correspondence between self-healing action sets and risk levels based on the fusion feature mean values; and a fourth determining submodule, used to determine the risk level of multiple self-healing actions in each self-healing action set based on the correspondence.

[0084] Optionally, the self-healing risk assessment device for the power distribution network further includes: a fifth acquisition unit, used to sort the risk levels of multiple self-healing actions in descending order after determining the risk level of the self-healing action based on the fusion feature vector, to obtain a first descending order sorting result; a second determination unit, used to determine the processing order for processing multiple risk events based on the first descending order sorting result, wherein the order of risk levels in the first descending order sorting result is positively correlated with the processing order; and a first control unit, used to control the execution of the self-healing action for processing each risk event in sequence according to the processing order.

[0085] Optionally, the self-healing risk assessment device for the power distribution network further includes: a sixth acquisition unit, used to sort the risk levels of multiple self-healing actions in descending order after determining the risk level of the self-healing action based on the fusion feature vector, to obtain a second descending order sorting result; a third determination unit, used to determine the execution order of multiple self-healing actions based on the second descending order sorting result, wherein the order of risk levels in the second descending order sorting result is positively correlated with the execution order; and a second control unit, used to control multiple self-healing actions to process multiple risk events sequentially according to the execution order.

[0086] According to another aspect of the present invention, a self-healing risk assessment system for a distribution network is also provided, wherein the self-healing risk assessment system for a distribution network uses any of the above-described methods for assessing the self-healing risk of a distribution network.

[0087] According to another aspect of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein the program executes any of the above-described methods for self-healing risk assessment of power distribution networks.

[0088] Optionally, in this embodiment, the computer-readable storage medium may be located in any computer terminal in a group of computer terminals in a computer network, or in any communication device in a group of communication devices.

[0089] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: acquiring multiple risk events of the distribution network, wherein a risk event is an event that causes a fault in the distribution network; inputting the multiple risk events into a self-healing model to process the risk events using the self-healing model, thereby obtaining multiple self-healing actions for processing the risk events, wherein the self-healing model is trained using multiple sets of first training data through deep learning, and each set of the multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event; acquiring the execution status, response speed, and impact range of the multiple self-healing actions, wherein the execution status refers to the incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range within which the risk event causes a fault in the distribution network; acquiring the fused feature vector of the execution status, response speed, and impact range in each self-healing action; and determining the risk level of each self-healing action based on the multiple fused feature vectors.

[0090] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: performing feature transformation on the execution state to obtain a first feature vector of the execution state; performing feature transformation on the response speed to obtain a second feature vector of the response speed; performing feature transformation on the influence range to obtain a third feature vector of the influence range; and fusing the first feature vector, the second feature vector, and the third feature vector to obtain a fused feature vector of the execution state, the response speed, and the influence range.

[0091] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: determining the risk level of a self-healing action based on a self-healing risk library and a fusion feature vector, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process historical risk events, and historical fusion feature vectors of historical self-healing actions; processing the fusion feature vector using a risk assessment model to obtain the risk level of the self-healing action, wherein the risk assessment model is trained using multiple sets of second training data through machine learning, and each set of multiple sets of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

[0092] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: acquiring multiple historical risk events of the distribution network within a historical time period; processing the multiple historical risk events using a self-healing model to obtain multiple historical self-healing actions for processing the multiple historical risk events; determining the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; and establishing a self-healing risk database based on the correspondence between the multiple historical risk events, the multiple historical self-healing actions, and the multiple historical fusion feature vectors.

[0093] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: comparing multiple fusion feature vectors with historical fusion feature vectors of historical self-healing actions in the self-healing risk database to obtain comparison results; determining multiple historical fusion feature vectors identical to the fusion feature vectors as target fusion feature vectors based on the comparison results; dividing multiple self-healing actions into multiple self-healing action sets based on the multiple target fusion feature vectors, and calculating the mean of the target fusion feature vectors of multiple self-healing actions in each self-healing action set to obtain multiple fusion feature mean values; determining multiple risk levels equal to the number of self-healing action sets; determining the correspondence between self-healing action sets and risk levels based on the fusion feature mean values; and determining the risk level of multiple self-healing actions in each self-healing action set based on the correspondence.

[0094] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: sorting the risk levels of multiple self-healing actions in descending order to obtain a first descending order sorting result; determining the processing order for processing multiple risk events based on the first descending order sorting result, wherein the order of risk levels in the first descending order sorting result is positively correlated with the processing order; and controlling the execution of self-healing actions for processing each risk event in sequence according to the processing order.

[0095] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for performing the following steps: sorting the risk levels of multiple self-healing actions in descending order to obtain a second descending order sorting result; determining the execution order of multiple self-healing actions based on the second descending order sorting result, wherein the order of risk levels in the second descending order sorting result is positively correlated with the execution order; and controlling multiple self-healing actions to process multiple risk events sequentially according to the execution order.

[0096] According to another aspect of the present invention, a processor is also provided, which is used to run a program, wherein the program executes any of the above-described self-healing risk assessment methods for power distribution networks.

[0097] According to another aspect of the present invention, a computer program product is also provided, including computer instructions, which, when executed by a processor, perform any of the above-described methods for self-healing risk assessment of power distribution networks.

[0098] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0099] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For instance, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0104] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for assessing the self-healing risk of a power distribution network, characterized in that, include: Multiple risk events of the distribution network are obtained, wherein the risk events are events that cause the distribution network to fail; Multiple risk events are input into a self-healing model to process the risk events and obtain multiple self-healing actions for processing the risk events. The self-healing model is trained using multiple sets of first training data through deep learning. Each set of the multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event. The execution status, response speed, and impact range of multiple self-healing actions are obtained, wherein the execution status refers to the degree of incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range of the distribution network failure caused by the risk event. Obtain the fused feature vector of the execution state, response speed, and influence range in each self-healing action; The risk level of each self-healing action is determined based on the multiple fused feature vectors.

2. The self-healing risk assessment method for distribution networks according to claim 1, characterized in that, Obtain the fused feature vector of the execution state, response speed, and influence range in each self-healing action, including: The execution state is subjected to feature transformation to obtain a first feature vector of the execution state; The response speed is subjected to feature transformation to obtain a second feature vector of the response speed; The influence range is subjected to feature transformation to obtain the third feature vector of the influence range; The first feature vector, the second feature vector, and the third feature vector are fused to obtain the fused feature vector representing the execution state, the response speed, and the influence range.

3. The self-healing risk assessment method for distribution networks according to claim 1, characterized in that, The risk level of each self-healing action is determined based on multiple fused feature vectors, including at least one of the following: The risk level of the self-healing action is determined based on the self-healing risk library and the fusion feature vector, wherein the self-healing risk library is a risk library established based on historical risk events, historical self-healing actions that process the historical risk events, and historical fusion feature vectors of the historical self-healing actions. The risk assessment model is used to process the fused feature vector to obtain the risk level of the self-healing action. The risk assessment model is trained using multiple sets of second training data through machine learning. Each set of second training data includes: a sample fusion feature vector and a sample risk level corresponding to the sample fusion feature vector.

4. The self-healing risk assessment method for distribution networks according to claim 3, characterized in that, Before determining the risk level of the self-healing action based on the self-healing risk library and the fused feature vector, the process includes: Obtain multiple historical risk events of the power distribution network within a historical time period; The self-healing model is used to process multiple historical risk events to obtain multiple historical self-healing actions for processing the multiple historical risk events; Determine the historical fusion feature vectors corresponding to the multiple historical self-healing actions respectively; The self-healing risk library is established based on the correspondence between multiple historical risk events, multiple historical self-healing actions, and multiple historical fusion feature vectors.

5. The self-healing risk assessment method for distribution networks according to claim 3, characterized in that, The risk level of the self-healing action is determined based on the self-healing risk library and the fused feature vector, including: The multiple fused feature vectors are compared with the historical fused feature vectors of the historical self-healing actions in the self-healing risk database to obtain the comparison results; Based on the comparison results, multiple historical fusion feature vectors that are identical to the fusion feature vector are identified as target fusion feature vectors. Based on the multiple target fusion feature vectors, the multiple self-healing actions are divided into multiple self-healing action sets, and the average value of the target fusion feature vectors of the multiple self-healing actions in each set of self-healing actions is calculated to obtain the average value of multiple fusion features. Determine a plurality of risk levels that are the same number as the set of self-healing actions; The correspondence between the self-healing action set and the risk level is determined based on the mean of the fusion features; The risk level of multiple self-healing actions in each set of self-healing actions is determined based on the correspondence.

6. The self-healing risk assessment method for distribution networks according to claim 1, characterized in that, After determining the risk level of the self-healing action based on the fused feature vector, the method further includes: The risk levels of the multiple self-healing actions are sorted in descending order to obtain the first descending order sorting result; The processing order for handling multiple risk events is determined based on the first descending sorting result, wherein the order of risk levels in the first descending sorting result is positively correlated with the processing order. The self-healing actions for each risk event are executed sequentially according to the processing order.

7. The self-healing risk assessment method for distribution networks according to claim 1, characterized in that, After determining the risk level of the self-healing action based on the fused feature vector, the method further includes: The risk levels of the multiple self-healing actions are sorted in descending order to obtain a second descending order sorting result; The execution order of the multiple self-healing actions is determined based on the second descending sorting result, and the order of risk levels in the second descending sorting result is positively correlated with the execution order; The multiple self-healing actions are controlled to process the multiple risk events sequentially according to the execution order.

8. A self-healing risk assessment device for a power distribution network, characterized in that, include: The first acquisition unit is used to acquire multiple risk events of the distribution network, wherein the risk events are events that cause the distribution network to fail; The second acquisition unit is used to input multiple risk events into the self-healing model, so as to process the risk events using the self-healing model and obtain multiple self-healing actions to process the risk events. The self-healing model is trained using multiple sets of first training data through deep learning. Each set of the multiple sets of first training data includes: a sample risk event and a sample self-healing action corresponding to the sample risk event. The third acquisition unit is used to acquire the execution status, response speed and impact range of multiple self-healing actions, wherein the execution status refers to the degree of incompleteness of the self-healing action, the response speed refers to the speed at which the distribution network executes the self-healing action, and the impact range refers to the range of the distribution network failure caused by the risk event. The fourth acquisition unit is used to acquire the fused feature vector of the execution state, the response speed and the influence range in each self-healing action; The first determining unit is used to determine the risk level of each of the self-healing actions based on the plurality of fused feature vectors.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein the program executes the self-healing risk assessment method for a distribution network as described in any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that, When the computer instructions are executed by the processor, they perform the self-healing risk assessment method for the power distribution network as described in any one of claims 1 to 7.

Images